Skip to main content

Full text of "Median Succumbs to Persian after Three Millennia of Coexistence: Language Shift in the Central Iranian Plateau"

See other formats

*-">»•* stmes 

BRILL Journal ofPersianate Studies 2 (2009) 62-87 

Median Succumbs to Persian after Three Millennia of 

Coexistence: Language Shift in the Central 

Iranian Plateau 

Habib Borjian 


The so-called Central Plateau Dialects or simply Central Dialects belong to the South Median 
group of Northwest Iranian languages and are spoken in central Iran, where the prevailing lan- 
guage is Persian. Currently, vestiges of these dialects are limited to several dozen remote villages 
as well as to the older generation of the Jewish and Zoroastrian communities living in the cities 
and in diaspora. The dominant influence of Persian for more than a millennium has resulted in 
the ousting of the vernaculars not only in major towns but also in a majority of villages. His- 
torical evidence suggests that Central Dialects were native to the entire central Iranian Plateau, 
larger towns included, until the late medieval period. The big shift may have taken place during 
and after the Safavid dynastic rule, perhaps as a result of forceful propagation of Shi'ism, among 
other economic and socio-political vicissitudes of those days. Concrete evidence becomes avail- 
able only in the later nineteenth century when European travelers and local geographers began 
to report on the language situation of the area. These documents enable us to speculate on the 
patterns and rates of language shift in various regions speaking Central Dialects. This trend has 
been accelerating parallel with the enormous socio-economic changes in the last half century. In 
many villages the local dialect is moribund and becoming increasingly limited to the elders, and 
the extinction will be the inevitable result of the forces of modernization and globalization in 
general and the rapid expansion of Persian education and mass media in particular. This paper 
attempts to show the dynamics of language shift among Central Dialects. The possible causes of 
the shift within village communities is discussed, while the urban Jewish and Zoroastrian speak- 
ers receive individual attention. Part of the data comes from the author's own fieldwork. 1 


endangered languages, language shift, Central Plateau Dialects, Jewish and Zoroastrian dialects, 
Iranian languages 

1 Preliminary versions of this paper were presented in the following conferences: Endangered 
and Minority Languages and Language Varieties, Georgetown University Roundtable on Languages 
and Linguistics, Washington, D.C., 3-5 March 2006; Second Iranian Linguistic Conference, Ham- 
burg, 17-19 Aug. 2007. This paper carries on my recent articles on the Central Dialects: "The 
geographical distribution of 'Provincial Dialects' in the district of Isfahan" (in Persian), Iranshe- 
nasi 17, 2005; and "Isfahan xx. Geography of the Median Dialects of Isfahan," Encyclopaedia 
IranicaXlV,-pp. 84-93. 

© Koninklijke Brill NV, Leiden, 2009 DOI: 10.1163/187471609X454671 

H. Borjian I Journal of Persianate Studies 2 (2009) 62-87 63 

Background: Defining Central Dialects 

Position among the languages of Iran and Iranian languages. The linguistic 
mosaic of Iran consists of Persian as the official language; the literary and semi- 
literary Turkish, Kurdish, Balochi, Turkmen, and Arabic, spread along and 
across the national borders; the non-literary Mazandarani, Gilaki, and Talysh 
along the southern shores of the Caspian Sea; and dozens of dialects spread 
throughout the interior of the country The last belong to the Western branch 
of Iranian languages and can be divided linguistically and topographically into 
two main groups of Southwest and Northwest. The Southwest dialects are 
centered in the southern province of Fars (the Persis of the Classical sources), 
hence usually referred to as Perside, the group which also embraces the Lori- 
Bakhtiari subgroup and Persian, once the lingua franca of the Persianate world. 
The Northwest Iranian languages consist of the aforementioned (1) Kurd- 
ish, (2) Balochi, and (3) the Caspian languages of Gilan, Mazandaran, and the 
Semnan region, (4 and 5) Gorani and Zaza within the Kurdophone areas, and, 
lastly, a series of dialects spread over a wide area in the northwest-central parts 
of Iran from Azerbaijan province to Isfahan and Yazd. This area is historically 
known as Media, thus the designation "Median" is proposed by certain schol- 
ars for these dialects (See below). Within these Median dialects, two subgroups 
are distinguishable: (6) the Tatic, consisting of Talysh and Tati-Azari, spoken 
along a belt from Lankoran on the western shore of the Caspian Sea all the 
way south to Qazvin, and (7) Central Dialects, the subject of this study, 
spreads over a large language area representing a wide variation in speech 
forms. There are also transition dialects, between the latter two subgroups, 
consisting of Vafsi, Alviri-Vidari, Tafreshi, Ashtiani, Amora'i, and Kahaki in 
the region between Hamadan, Sava, and Qom. 2 

Geographical distribution. The domain of Central Dialects is clearly defined by 
the Central Persian Desert in the east and the massive Zagros range in the 
west, by and large within the modern provinces of Isfahan and Yazd. On the 
foothills of the Zagros lie, clockwise from the south, Khunsar, Golpayegan, 
Khomeyn, and Mahallat — historically secluded valleys relatively distant from 
major highways. The Karkas range, an offshoot of the Zagros, begins just east 
of Mahallat and west of Kashan and stretches southeastward some 200 miles, 
surrounded by villages speaking Central Dialects. On the west side of the 
Karkas, along the Qom-Isfahan highway, lie Delijan and Meyma-Jowshaqan 
districts, both connected to the districts of Kashan and Natanz across the 

Cf. Windfuhr, 1989; Stilo, "A Description. 

64 H. Borjian I Journal of Persianate Studies 2 (2009) 62-87 

range by a network of trails and roads leading to many secluded Median- 
speaking villages. Running along the edge of the desert, the Qom-Kashan- 
Yazd highway traverses the Median-speaking districts of Ardestan and Na in, 
and to a lesser degree Ardakan and Yazd. These thinly-inhabited steppe-desert 
districts are separated from Isfahan plain by an extension of the Karkas range, 
where the mountain chain lowers and spreads out, sometimes called Marshnan 
after its highest peak. This section of the range is dotted on both flanks with 
villages speaking closely-related dialects, including those of Kuhpaya, a size- 
able sub-district of Isfahan. Finally, the plain of Isfahan, irrigated by the 
Zayandarud, constitutes the southern boundary of the Median-speaking prov- 
ince. The grouping of Median dialects in Isfahan proper is best achieved within 
the well-defined economic units traditionally called boluks, namely Marbin, 
Borkh w ar, Jarquya, Rudasht, and Kuhpaya. To the east of Isfahan is the prov- 
ince of Yazd with Median-speaking Zoroastrian and Jewish communities. 3 

A few Median-speaking localities are geographically isolated from the con- 
tinuum of the Central Dialect area. These include the town of Semnan on the 
Tehran-Khorasan highway, and the village Sivand north of Persepolis, repre- 
senting a migratory dislocation southward, perhaps in medieval times. Another 
distant outlier is the southeastern city of Kerman, with Zoroastrian and Jewish 
residents speaking vernaculars remarkably similar to the Median of Yazd. The 
fact that the Jewish community of Kerman emigrated from Yazd in the nine- 
teenth century and that Zoroastrians of both cities had close ties over centu- 
ries, all suggest that the linguistic flow must have been from Yazd to Kerman, 
and that the presence of Median in Kerman should be quite recent. 4 Lastly, 
aberrant forms of Central Dialects spoken by the Jewish communities of 
Hamadan and nearby towns mark the northwestern corner of Central Dia- 
lects' domain. 

Speakers and domain of use. The speakers of Central Dialects live in more than 
200 villages and in a few towns where Persian is the primary means of com- 
munication. 5 These speech communities range in size from a few to several 
hundred households, but almost never more than 10,000 speakers. 6 It would 
be unrealistic to calculate the number of speakers because the users of the 
dialects are increasingly shrinking to older generations. In larger towns such as 

3 For detailed geographical grouping of these districts, see H. Borjian, 2007d. 

4 Thus the designation "Kermanian" for Central Dialects, recently used by Pierre Lecoq 
(2002, p. 1), is untenable. See also the discussion in my review of this book in Studia Iranica 34, 
2005, pp. 306-308. 

5 For the economic geography of the Central Dialect area, see, H. Borjian, 2007c. 

6 H. Borjian, 2005b; idem, 2007d. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 65 

Na'in and Ardestan, Median has been limited to the less affluent quarters or, 
as in Isfahan, to the fading Jewish communities. The domain of use varies. In 
the south, where Persian permeation is less advanced, native dialects are used 
for most in-group communication among the residents, while Persian or the 
local dialect may be used for inter-village communication, depending on the 
degree of intelligibility. 7 In the north and in larger towns there is a high degree 
of use of Persian in all but the most intimate communication. 8 A great major- 
ity, if not virtually all, of the speakers are now diglossic in their mother tongues 
and Persian. 

Names of the dialects. Scholars of Iranian languages have generally agreed on 
the designation "Central Dialects" ever since its coinage at the turn of the 
twentieth century by Wilhem Geiger in Grundriss der Iranischen Philologie 
(Strasbourg, 1898-1901). A viable alternative would be "Median," or "New 
Median," a term initiated by Clement Huart and later revived by Ehsan Yar- 
shater. 9 It has the advantage of not only giving a historical dimension both to 
the language and its provenance (see Historical Perspective below), but also the 
convenience the adjective "Median" offers (e.g. Median-speaking villages, he 
speaks Median, etc.) but "Central Dialects" lacks. To distinguish Central 
Dialects from the Tatic group, one may call them South Median and North 
Median, respectively. 

Each Median dialect is generally named after the community of its speak- 
ers; that is usually the village or town and sometimes the district where the 
dialect is spoken. The residents of the township of Kupa, east of Isfahan, call 
their own vernacular virze 'belonging to Vir (the local and historical name of 
Kupa)', and they refer to the language of the neighboring village of Qehi as 
kize. Likewise, the Jewish community of Isfahan calls its dialect jidi 'Jewish.' 
The dialects of the Zoroastrians are usually referred to as gabri by outsiders and 
dari by the community itself. Speakers of the northern vernaculars of Central 
Dialects in Kashan province identify their shared language either as dei 'of vil- 
lage' or as raji (or rayeji), 10 which implies a distant reminiscence of the city of 
Raga/Ray, thus relating the language of this area to the now extinct idioms 
around Tehran. As one travels southward towards Isfahan, the term velati 
(a short form (Aveldyati 'provincial') becomes prevalent. A curious term com- 
monly used to distinguish the dialects is buro-beshe, for in many Central Dia- 
lects buro and beshe are the typical imperatives of 'come' and 'go,' respectively. 

7 Krahnke, 1976, p. 58. 

8 Cf. Majidi, 1975, passim. 

5 Yarshater, 1971, p. 17; idem, "Azerbaijan vii." 
10 Yarshater, 1985. 

66 H. Borjian /Journal ofPersianate Studies 2 (2009) 62-87 

In turn, the villagers of Abchuya, south of Na'in, designate their dialect as 
osme-siga, literally 'now — thus,' two words typical of the Na'ini subgroup of 
these dialects. 11 The inhabitants of the plain of Rudasht refer to the dialect of 
their neighboring mountaineers of Kuhpaya as osme-uso 'now — go!', contrast- 
ing with their own zonon 'now' and beshe 'go!' 12 These distinctions made by 
native speakers reflect the recognition of certain isoglottic divisions that sepa- 
rate Central Dialects. 

Internal divisions. The grouping of Central Dialects has presented substantial 
problems because of numerous intersecting isoglosses and the lack of clear 
isoglottic bundling, a pattern which is probably the result of repeated internal 
migrations. Examples of crossing isoglosses are the words for 'dog' kua - espa; 
'thus' nezan - issin - sige; 'now' baton - zogun - zonun - osme; 'large' gord - 
bele - mas - gozdr. Several important morphological traits, such as the exis- 
tence or lack of grammatical gender, 13 differentiate the dialects. Comparativ- 
ists have tried to construct a reasonable grouping of Central Dialects based on 
the historical development of their sound systems as well as a selective number 
of morphological and syntactic features. The most widely accepted classifica- 
tion, however, is the quadratic system of NW, NE, SW, and SE, which is 
indeed more of a topographical than strictly linguistic division. At any rate, no 
neat demarcation lines between groups can be drawn. 14 

As to the number of languages in the Central Dialect area, no conclusion 
can be drawn from the available comparative studies. A governing criterion 
can be the level of mutual intelligibility, which varies among the subdialects, 
usually attenuating with distance, and occasionally abruptly. No detailed study 
is available on the issue of comprehensibility among subdialects, 15 but a recent 
survey within the district of Isfahan by the present author maintains that the 
traditional economic units, called boluks, essentially demarcate the mutually 
intelligible groups from their neighbors. 16 Thus in Isfahan district alone we 
may assume the existence of about ten South Median languages, the number 
of subdialects of each ranging from one (for Sedehi, spoken in the Marbin 
plain to the west of Isfahan, and Jidi, the Median dialect of the Jewry of 
Isfahan) to dozens (for Rudashti and Kupayi). 

11 Krahnke, 1976, pp. 55-56, 263. 

12 Author's field notes, taken in the village Seryon in Rudasht in 2004. 

13 See Yarshater, 1985. 

14 See Yarshater, 1985, esp. pp. 743ff.; Lecoq, 1989; Windfuhr, "Central Dialects"; Stilo, 
2007. For a history of classifications, see H. Borjian, 2005b. 

15 Cf. Krahnke, pp. 58-59. 

16 H. Borjian, 2005b. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 67 

It may be fitting to note that only two of these ten languages, namely Gazi 
and Jidi, are listed in Ethnologue. 17 Among the 75 living languages Ethnologue 
lists for Iran, nine belong to the Central Dialect group: Dari (spoken by the 
Zoroastrians of Yazd), Dzhidi (sicl i.e. Jidi, the Jewish dialect of Isfahan), 18 
Gazi, Khunsari, Natanzi, Na'ini, Semnani, Sivandi, and So(h)i. 19 This list fails 
to identify the true distribution of the representative languages among Central 
Dialects; it rather lists without any comparative analysis only the dialects com- 
monly known to linguistics worldwide. 

Documentation and Study of the Dialects 

The documentation of Central Dialects began in the first half of the nine- 
teenth century, but continued more systematically during 1880-1940, and 
then, after the politically-troubled years that followed World War II, during 
the 1960s- 1970s. The Revolution of 1979 denied foreigners of fieldwork, but 
collections by local enthusiasts and students has since been on the rise. 

Among the earlier reports, the pride of place belongs to Valentin Zhu- 
kovskii's Material)/, 10 embracing the dialects of Vanishan, Qohrud, Kesha, 
Zefra, Gaz, Sedeh, Kafran, Sivand, and the Jewry of Kashan. Oskar Mann 
carried out fieldwork during the first decade of the twentieth century to much 
of the Central Dialect area, collecting data from Mahallat, Khunsar, Natanz, 
Soh, and, particularly extensively, from Na'in, as well as from Sivand and Sem- 
nan. His data was edited and published posthumously by Karl Hadank. 21 The 
primary collectors of the dialect of Zoroastrians of Yazd and Kerman are Major 

17 Ethnologue: Languages of the World is a web and print publication of SIL International, a 
Christian linguistic service organization which studies lesser-known languages primarily to pro- 
vide the speakers with Bibles in their native language. Its latest edition (2005) contains statistics 
for 6,912 languages and gives the number of speakers, location, linguistic affiliations, etc. It is 
currently the most comprehensive existing language inventory, along with the Linguasphere 
Register. However, the accuracy of data in the Ethnologue for many languages has been disputed 
by various scholars. This author has found the Ethnologue quite inaccurate in every respect for 
the languages spoken in Iran; for instance, the speakers of Persian and Azerbaijani Turkish lan- 
guages in Iran are estimated as 36 and 37 percent, respectively; the invariable population figure 
of 7,033 is mentioned for the speakers of Gazi, Na'ini, Natanzi, and Sivandi. See http://www. 

18 It is incorrectly interpreted in Ethlologue as Judeo-Persian, while the latter refers commonly 
to the forms of Persian written in Hebrew script. 


20 Zhukovskii, 1888-1922. 

21 Mann and Hadank, 1926. 

68 H. Borjian /Journal ofPersianate Studies 2 (2009) 62-87 

D.L.R. Lorimer 22 and Vladimir Ivanow; 23 the latter's works also cover the dia- 
lects of Anarak, Khur and Mehrajan along the edge of Central Desert. Arthur 
Christensen's fieldwork, 24 covering Natanzi, Farizandi, and Yarandi, remains 
the only one of its kind for the latter two idioms. The briefer works are those 
of Ruben Abrahamian 25 on the Jewish dialects of Isfahan and Hamadan, Sir 
Harold Bailey 26 on Ardestani, and Ann K.S. Lambton 27 on Meyma'i and Jow- 
shaqani. Wilhelm Eilers' extensive texts from Gaz, Khunsar, and Sivand, col- 
lected from 1936-40 and then in the 1960s, 28 render these idioms the most 
comprehensively documented of all Central Dialects. 

A fresh beginning was made in the late 1 960s, when Pierre Lecoq conducted 
fieldwork on the dialects of Qohrud, Abuzeydabad, Abyana, Tar, Badrud, 
Na'in, Anarak, Ardestan, Varzana, and Sivand. 29 Karl John Krahnke visited 
sixteen villages in 1 970-72 within the triangle of Natanz-Na'in- Isfahan, con- 
centrating on the vernaculars of Nohuj, Na'in, Natanz, and Ardestan. His 
documentation, however, remains unpublished, save for the relevant materials 
he used in his unique comparative study 30 Studies of Ehsan Yarshater and 
Donald Stilo on Central Dialects appear under various entries in Encyclopae- 
dia Iranica. The engagement of the Persian Academy of Letters (Farhangestan) 
in the study of dialects in the 1 970s yielded several short but useful studies, 
including Majidi's survey, 31 with sample glosses from the last generations of 
speakers in the Kashan-Mahallat-Meyma area; Bahram Farahvashi's Khuri 
glossary 32 and Haideh Sahim's masters' thesis on the Hamadani Jewish, 33 
among others. Postgraduate theses have been growing exponentially in the last 
two decades, yet many suffer from inaccuracy and methodological problems. 
To these, one may add publication of glossaries, idioms, and poems by native 
pundits on their home villages. 34 

22 Lorimer, 1916; idem, 1928; Vahman and Asatrian, 2002. 

23 Ivanow, 1926; idem, 1927; idem, 1934-39. 

24 Christensen, 1930. 

25 Abrahamian, 1930. 

26 Bailey, 1935. 

27 Lambton, 1938. 

28 Eilers and Schapka, 1976-88. 
25 Lecoq, 1979. 

30 Krahnke, 1976. 

31 Majidi, 1975. 

32 Farahvashi, 1976. 

33 Note also a comparative study of Jewish dialects of Hamadan, Isfahan, and Shiraz in Sahim, 

34 For a comprehensive bibliography of documentations, see Windfuhr, "Central Dialects"; 
H. Borjian, 2007d. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 69 

Two related nationwide projects have been conceived towards generating a 
linguistic atlas for Iran based on audio-recorded materials. The first one began 
in 1974 at Farhangestan to identify dialects nationwide based on a question- 
naire consisting of 150 glosses and twenty sentences. 35 As the revolution 
soon interrupted the project, the collected data were transferred to the Cul- 
tural Heritage Foundation, which has published four volumes, all on Isfahan 
province. 36 The second project, sponsored by the Agriculture Corps (Jehad-e 
keshdvarzi) since 1998, is by and large a replication of the former project: it 
incorporates similar strategy, techniques, and questionnaires, except that the 
number of sentences is now doubled. The greater part of the survey had been 
completed in 2005, when I visited their Tehran headquarters; however, the 
poor quality of workmanship and audio-recordings limit their merit to the 
mere identification of the language in each surveyed locality, when the audio- 
cassettes are not lost or field notes are legible. 

The published linguistic materials, substantial as they may appear, consti- 
tute a tiny fraction of the documentation needed before Central Dialects van- 
ish altogether. Most of the community speeches remain unknown, and an 
inventory of the localities in which Central Dialects are spoken is yet unavail- 
able. Documentations of quality are indeed few and are limited largely to 
those of the earlier efforts made by the scholars of Iranian philology. The main 
thrust of scholarly interest in these dialects has been comparative, with the 
intention of clarifying areal or genetic relationships. There have as yet been 
relatively few attempts to synchronize the work of traditional disciplines with 
the global enterprise of language documentation. As to the socio-linguistics of 
the area, the only study known to me is the brief sketch Karl Krahnke included 
in an introductory chapter of his otherwise comparative dialectology. 37 

Historical Perspective 

The provenance of modern Iranian languages corresponds to a large extent to 
areas where Middle and Old Iranian languages were once spoken. Among the 
living Iranian languages, Persian is the only one with an established history 
from its ancient to its modern form; historical evidence is insufficient to estab- 
lish as close a relation between the various chronological stages of other 

35 Thamara, 2002. 

36 Shendsdyi-e guyeshhd-ye Iran (1983-2000), which has published only a fraction of the col- 
lected data: 1 9 glosses and two short sentences for two representative dialects in each rural dis- 

37 Krahnke, 1976, pp. 14-20. 

70 H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 

Iranian languages. 38 The geographical position, however, may help us to arrive 
at certain conclusions regarding Central Dialects. 

Ancient Media and the Median. The earliest known language spoken in the 
Central Dialect area is Median, the language of the Medes who migrated into 
the Iranian Plateau, as the Persians did in the early first millennium B.C.E. 
Median is generally assumed to be the official language during the dynastic 
rule of the Medes (ca. 700-559 B.C.E.), but no text has been discovered in 
Median; only numerous non-Persian words in the Old Persian texts are com- 
monly presumed to be Median, while other Median forms are preserved in the 
Akkadian versions of the Achaemenid inscriptions and in classical works 39 — 
for instance, Herodotus (1.110) cites the Median word for 'dog' as spaka, 
various forms of which are current in many Central Dialects of our time. 40 
Subsequently, Old Median, as we know it today, is little more than a set of 
reconstructed words that clearly mark its Northwest Iranian characteristics, 
contrasting with Old Persian that belongs to the Southwest group. This lin- 
guistic evidence alone is inadequate to link Median with any specific later 
dialect or group of dialects of the same language family 41 As there are no solid 
isoglottic features that distinguish Median from proto-Parthian, the linguistic 
interpretation of Old Median (already with many dialects) can be made as 
broad as to consider it an ancestor to Parthian and all other Northwest Iranian 
languages, including Central Dialects, the Tatic and Caspian groups, Gorani- 
Awromani, and Zaza. 42 

While linguistic analysis alone is insufficient to hold up the designation 
"New Median" for Central Dialects (and the Tatic group as well), the geo- 
graphical factors neatly support such designation: (1) The facts that no mass 
migration was ever reported into the Median territories after that of the Medes 
and that there is no historical evidence that the population of Media was ever 
dislocated on a significant scale, suggest that Central Dialects and Tatic should 
have ultimately descended from ancient varieties of Median, and (2) no major 
ancient or medieval inscriptions have been found in the Central Dialect area 
to suggest that any other major Iranian language has ever been spoken there 

38 For an overview, see Skjaervo. 

39 See Schmitt, esp. pp. 86-90. 

40 Median spa-ka- compares with Old Persian sa-ka- > New Persian sag 'dog'. It typifies one of 
the earliest isoglottic splits (sp - s) that separates Persian and other Perside languages from the rest 
of languages of Iranian stock. 

41 Gemot L. Windfuhr, review of Yarshater, 1971, in JAOS 92, 1972, pp. 370-374; Krahnke, 
1976, p. 51. 

42 Kurdish and Baluchi, however, do not belong to the Median group, for they carry Perside 
elemets in their oldest phonological strata. See MacKenzie; Windfuhr, 1975. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 7 1 

by a sizable community. Even if we choose to ignore language-related argu- 
ments altogether, the mere fact that the area covered by Central Dialects and 
Tatic corresponds roughly to the old Greater and Lesser Media, respectively, 
provides sufficient grounds to use the designation "New Median" for these 
dialects. 43 

Medieval Fahlavi survivals. The earliest solid linguistic evidence in the Central 
Dialect area surfaces in medieval times. The Islamic works written between the 
tenth and fifteenth centuries have several citations of words and phrases from 
the native dialects of the area that correlate closely to the living Central Dia- 
lects. Commonly referred to as "Pahlavi" or "Fahlavi" 44 (i.e. Parthian), these 
medieval dialects had an oral literary tradition, with specimens quoted in 
some anthologies of Persian poems. The fact that Fahlavi dialects, as the 
continuation of Old Median dialects, have certain linguistic affinities with 
Parthian suggests a good deal of language mixture with Parthian, another 
Northwest Iranian language which belonged originally to the super-province 
of Parthia, the present Khorasan, but was gradually spread westward into 
Media. 45 Another major source of influence on the Central Dialects has been 
Persian, the lingua franca of the region for a millennium. Persian is not only a 
distant maternal cousin of Central Dialects (as is the case for any other Iranian 
language spoken on the Plateau) but in a sense their father as well: it has 
increasingly been flooding the dialects, primarily their vocabulary and, more 
recently, permeating their morphological and syntactical structure. 

The Fahlavi survivals reveal important facts about the language of the urban 
centers that are now entirely Persophone. Quotations in two eleventh-century 
works from the vernacular of Isfahan and its environs 46 leave little doubt that 
the city had not yet fully adopted Persian by that time, even if it was to become 
the hub of Persian literature in the centuries to come. In fact, Fahlavi must 
have been spoken in the city as late as the fourteenth century, for we find three 
poems in the native speech of Isfahan in the collected works of Owhadi (1274- 
1338), the prominent Persian poet from Isfahan. 47 There are similar Fahlavi 
survivals from Kashan, another Persianized city of the Central Dialect area, 

43 Please note that claims of some kind of connection between the ancient Media and Median 
language on the one hand and Kutdistan and the modern Kurdish dialects on the other, raised 
by Kurdish pundits, have been refuted by D.N. MacKenzie; Windfuhr, 1975. 

44 Clement Huart called these dialects "Muslim Pahlavi" to distinguish it from Zoroastrian 
Pahlavi — that is Middle Persian. For a discussion, see H. Borjian, 2008b. 

45 For this reason these languages may properly be called Median-Parthian; see H. Borjian, 

46 Tafazzoli, 1971. 

47 AdibTusi, 1963; Borjian, 2009. 

72 H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 

whose old vernacular survived through its Jewry until a few decades ago. 48 
Therefore, it is safe to consider Central Dialects as being native to the entire 
area, larger towns included, until late medieval times. The fact that the south- 
ern and southwestern parts of Isfahan province show no survival of Median is 
due to the dominance in these regions, before the spread of Persian, of other 
languages, chiefly of the Perside Lori-Bakhtiari family. 

Safavid rule and thereafter. It is difficult to establish when and how the larger 
part of the population of the Central Dialect area shifted to Persian. A late 
nineteenth-century report 49 on the localities around Isfahan where "provin- 
cial" dialects were spoken shows a general picture not radically different from 
the areal pattern of the dialects today. Therefore, the lion's share of linguistic 
integration must have occurred before the nineteenth century but not earlier 
than the fourteenth century, when Owhadi wrote his dialect poems. A con- 
tributing factor to language shift here might have been the political status of 
Isfahan as the capital of Safavid Persia from ca. 1600-1722, during which time 
a rather abrupt shift to Persian could have taken place in the city, with subse- 
quent radiation outward in all directions, overlying the native dialects up to 
a radius of 50 km or so. 

It is indeed not difficult to explain how the shift came about in Isfahan 
when we consider the metropolitan status of the Safavid capital, which attracted 
bureaucrats, craftsmen, religious leaders, and merchants not only from through- 
out Persia but also from virtually the entire civilized world, so much so that 
immigrants outnumbered the native townspeople at least tenfold. 50 How 
could this enormous brassage not leave behind Persian as the language of the 

48 Yarshater, 1974. 

49 Janab, 1992, p. 128. 

^° Jean Chardin, who lived for ten years in the Safavid capital, offers, in his Voyages, a detailed 
and vivid description of "the greatest and most beautiful town of the whole Orient," in which 
there were to be found followers of all religions and merchants from the whole world. Indians 
involved in the banking business alone numbered no fewer than ten thousand. The commercial 
character of the Safavid metropolis can be seen in other figures offered by Chardin: there were 
1 62 mosques and 1 ,802 caravanserais within the city walls and 500 more caravanserais without. 
Accordingly, for every mosque there were more than eleven inns, a ratio that has become reversed 
today: 73 inns of all ranks vs. more than 1,500 mosques in the district of Isfahan (2003 data). 
Moreover, according to Chardin, there were twelve to fourteen thousand prostitutes publicly 
registered with as many more who worked on their own. And this was in the same city that was 
the capital of Shi'ism, attracting religious scholars from all Muslim lands; the Jabal 'Amel of 
Lebanon alone was the provenance of virtually all leading jurists of Isfahan in those days. The 
population of this enormous city, estimated by Chardin to match that of London (i.e. circa 
650,000), grew to well over one million, according to other reports, by the end of Safavid rule. 
See Ferrier, tr. and ed., 1996, passim; H. Borjian, 1993; idem, 2007a. 

H. Borjian I Journal of Persianate Studies 2 (2009) 62-87 73 

town? The countryside too, the source of bread for the city, could not have 
escaped from language shift, at least partially. There is indeed a clear link 
between the agricultural districts included in the water regulation system 51 
and those which shifted to Persian in the upper and middle course of the river 
Zayandarud. The topographic patterns of the shift may also be linked to the 
existence of the elaborate underground canals known as qandt, which usually 
required state investment and sponsorship; thus the boluk of BorkrTar which 
received its water exclusively from qandts is left with no more than three 
Median villages, while over seventy percent of the settlements in the rainfed 
boluk of Kuhpaya have preserved their dialects. 

The downfall of the Safavid state by the Afghan invasion of Isfahan in 1722 
was followed by the most dramatic demographic decline in the history of the 
city and its environs. The invasion itself and the consequent famine and 
destruction of the rural infrastructure resulted in the dwindling of the popula- 
tion from over a million souls just before the invasion to some 50,000 in 
1796, i.e. a population decline of over twentyfold in less than a century. 52 In 
the countryside there is evidence of population shifts in Jarquya district, 
located southeast of Isfahan: the population replacement in the village of 
Malvajerd following the Afghan invasion, and the forced relocation of Perso- 
phone tribes from Fars under the Zand dynastic rule in the late eighteenth 
century 53 could have resulted in language shift. Even more evidence such as 
this may surface when historical documents are carefully scrutinized. 

Did religion have a share in the language shift? Why in the bigger towns, 
such as Isfahan, Kashan, and Yazd, are the native dialects spoken exclusively 
by the Jewish and Zoroastrian communities, while all Shi'ite residents have 
shifted to Persian? Does this have anything to do with Persian being the means 
of intensive religious propaganda carried out under the Safavid dynastic rule? 
As is well-known, the language of the pulpit and its associated preaching styles, 
the genres of rowza-kh" 'dni and marthiya, have been well developed, taught, 
and practiced only in formal Persian — at least during the last couple of centu- 
ries in central Persia. More facts are needed before one can draw conclusions 
in this regard, but the religious factor appears to be the main cause of language 
shift in other large cities of Iran, such as Shiraz, which has abandoned its old 
Perside vernacular in favor of Persian, and Tabriz, which is entirely Turkish. 

Another venue to look into is the communication patterns attached to 
trade, clearly the route of much Persian influence on the native idioms. Whether 

51 The allocation of the Zayandarud's water to various districts around the city of Isfahan is 
documented in the Sheikh Baha'i's tumdr, dated 1527. See Lambton, 1937-39. 

52 See H. Borjian, 1993. 

53 Shafi'i Nikabadi, 1997, pp. 150, 446ff.; H. Borjian, 2005b. 

1\ H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 

urban peddlers traversed the countryside or villagers visited the urban market 
centers, the means of communication was Persian, and the bdzdri language 
could easily and quickly flood the dialects. 54 Neither must one overlook the 
carpet industry, which boomed in the 1870s in response to the growing mar- 
kets in the West, and became a significant source of income for rural house- 
holds in much of the Central Dialect area. This brought most villages into 
contact with the towns on an unprecedented scale. 55 

Rates and Patterns of the Shift 

Assuming that the entire central Persia once spoke Median dialects, let us first 
see what parts thereof have been more successful in resisting Persian. In terms 
of the number of settlements, the center of gravity is east of Isfahan in the 
Rudasht-Kuhpaya area. The most resilient of all appears to be Rudasht, with 
an uninterrupted continuum of nearly 50 Median-speaking villages. 56 In the 
neighboring Kuhpaya-Zefra-Sagzi area, in the southern foothills of Marshnan, 
more than three-quarters of the total 1 37 settlements are reported to have 
maintained their native speech, while only about half of Jarquya's 23 villages 
have resisted Persian. Outside of Isfahan district, the Meyma-Jowshaqan area 
yields the highest ratio: 19 Median-speaking localities of the total 23. The 
ratio in the rest of Kashan district is no more than a quarter of the 73 settle- 
ments. 57 In the northern and western parts of the province, Median survives 
in not more than a few isolated communities. Thus, the language shift seems 
to be more intensive as one travels from the SE to NW. This pattern can be 
attributed partly to the fact that the northern areas have been in closer contact 
with the capital city of Tehran, around which the Tatic dialects used to be 
spoken, but are now totally extinct. The Zagros districts of Golpayegan, Khun- 
sar, and Faridan have been under the influence of both Lori and Turkic. 

Persian permeation has been far more advanced in larger towns. Isfahan in 
the south and Kashan in the north, the largest urban centers of the area, have 
lost their native vernaculars for centuries, with the exception of their Jewish 
residents. Of medium-sized towns Golpayegan turned to Persian long ago, but 
the nearby Khunsar was reported in 1988 as having retained its native dialect 

54 Ivanow, 1934-39, p. 38. 

55 See, i.a., H. Borjian, 2007b. 

56 H. Borjian, 2007d. 

5 The statistics are calculated based on reports on individual areas in Shendsdyi-e guyeshha- 
ye Iran. 

H. Borjian I Journal of Persianate Studies 2 (2009) 62-87 75 

in certain quarters. 58 The incomplete shift in Khunsar is possibly because the 
town lost much of its former importance in the second half of the twentieth 
century, causing a temporary deceleration in language shift. Na'in appears to 
have had a similar fate, as Median was restricted to the older and lower-class 
residents in the early 1970s. 59 Natanz, another small town, was rapidly becom- 
ing Persophone already in the early 1930s and had lost much of its Median 
characteristics forty years later. 60 It appears that as a town grows in terms of 
population and urbanism, its native dialect becomes more vulnerable. 

Language maintenance in villages depends on the geographic location and 
arrangement. Those located in the vicinity of cities have been far more vulner- 
able to language loss. This trend is well attested east of Isfahan in the sub- 
district of Rudasht, where the Persian-Median frontier shifted a few villages 
further eastward from 1977 61 to 2004, when I visited the area. 62 A parallel 
trend is observed in Kashan district. 63 The proximity to major roads 64 has also 
been instrumental to language shift; a salient example, on the modern Isfahan- 
Tehran highway, is the village Murchakh w ort, where the local dialect was dying 
out in the 1970s. 65 In Ardestan district the rural communities along the 
Kashan- Yazd highway on the desert plain are almost entirely Persianized, con- 
trasting with those in higher valleys. For such areas as the piedmont district of 
Kuhpaya, which exhibits a pattern of Persianization along certain mountain 
valleys, further field investigation is needed, as migration into this remote 
district is unlikely. Finally, the shift to Persian seems to be far more advanced 
in isolated villages than those forming interconnected clusters, as in Rudasht, 
Zefra, and Jowshaqan. 

Role of Ethnic and Social Identities 

Ethnic relevance. The level of endangerment of languages and dialects of Iran 
appears to be correlated with ethnic or tribal identity of their speakers. The 
Kurds and the Baloch show a strong sense of ethnic identity as well as admira- 
tion for their mother tongue. These ethnolinguistic groups as well as the 
Turkish and Arabic speakers of Iran find additional political expression across 

58 Shendsdyi III, pp. 64-65. 

59 Krahnke, 1976, p. 58. 

60 Krahnke, 1976, pp. 77, 112. 
" Shendsdyi ill, pp. 117-21. 

62 H. Borjian, 2007d. 

63 Majidi, passim. 

64 Siroux, Chapter 1. 

65 Comparing the reports by Majidi, p. 13, and Shendsdyi II, p. 36. 

76 H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 

national borders. The Lors and Bakhtiaris of the southern Zagros are proud 
of their tribal affiliations, and their dialects appear less vulnerable to Persian, 
though concrete facts are yet to be collected. Towards the opposite end of the 
spectrum, the Caspian languages of Gilan, Mazandaran, and Semnan exhibit 
a high level of Persianism, partly due to a low sense of ethnic identity on the 
part of their millions of speakers. 66 As for the speakers of Central Dialects, the 
outstanding fact is that they stand on the lowest step of the identity ladder, 
lacking even the mere ethnolinguistic title the speakers of Tatic possess due to 
their presence within or near a Turcophone milieu. 

In fact, the population of the Central Dialect area is ethnically undifferenti- 
ated. There is no indication of any historical or contemporary tribal organiza- 
tion or affiliation. The language of a village or town, whether Median or 
Persian, plays no role in its identification, neither by the inhabitants nor by 
outsiders; an individual outside his hometown is usually identified with his 
provenance rather than his mother tongue. Nor is religion a differentiating 
factor (with the obvious exception of the Jews and Zoroastrians; see below), 
as the speakers of Central Dialects not only confess to Shi'ism, the state reli- 
gion, but the area has been known historically as the heartland of Shi'ism in 
Persia. Interestingly, many contemporary Grand Ayatollahs carry the top- 
onyms of the Central Dialect area in their surnames: Khomeini, Mahallati, 
Golpayegani, Khunsari, Kashani, etc. Moreover, no sign of regionalism has 
ever been recorded in the Central Dialect area. "In general, the people speak 
only of family, village, and national values and interests, recognizing only the 
political and administrative realities of the modern nation above and beyond 
those of the village." 67 The above is not too surprising given the fact that in the 
past half millennium central Iran has been the core of what is generally under- 
stood as Persian culture, and its population is the most conservative guardians 
of national traditions and, therefore, zealous antagonists to modernization. 

Prestige. In this respect, Central Dialects are probably no more disadvantaged 
than other languages and dialects spoken in Iran, where Persian is regarded as 
the language. This is not the place to talk about the invincible assertion of 
Persian throughout the Persianate societies in the past millennium; suffice to 
mention that forms of literary expression in the region has been almost exclu- 
sively in Persian, and that literary Persian is generally understood as the only 
form worthy of being written, taught, and preserved. Thus, it is common 
practice on the part of local collectors of dialects to justify their curiosity by 

66 For a sociopolitical discussion on the Mazandarani language, see M. Borjian, 2005, 
pp. 65-74. 

67 Krahnke, 1976, p. 20. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 77 

stating that their works may eventually contribute to the richness of Persian 
vocabulary by means of revitalizing those words or purer varieties that have 
been forgotten in Persian but are retained in the dialects. Those who write 
poetry in Central Dialects usually aim to cast their poems into the exact meters 
of classical Persian prosody even with partial success. 68 It is therefore not sur- 
prising that dialect literature tends to resort to a kind of tawdry vulgarity far 
less commonly found in formal Persian literature. 69 Isolated remarks recorded 
by past collectors of Central Dialects show predominantly negative attitudes 
held by speakers of the dialects toward their mother tongue, though they range 
from the belief that a given language is punishment given by God, to a feeling 
that the language was a precious inheritance allowing for broader communi- 
cational opportunities. 70 The fact that Central Dialects are shared with the 
urban Jewish community makes some villagers in the vicinity of Isfahan sup- 
pose that they may have a Jewish lineage — something they are not particularly 
proud of. The fact that some speakers of Central Dialects, usually more edu- 
cated ones, hold the conviction that their "Pahlavi" language is a purer and 
older form of Persian does not alter the balance of prestige that is rooted in 
social realities. 

There is also a socio-economic aspect to the question of esteem. Being spo- 
ken by villagers or the lower-class town dwellers, Central Dialects are seen as 
the low-prestige variety. The dialects are disdained by the younger generation 
and practically shunned by school-aged children. My personal observation 
also suggests that younger women hold a more negative view of their native 
idiom than male speakers generally do. Villagers of the central Persia are gen- 
erally "penurious, credulous and satisfied," 71 with perhaps one exception — 
those of Abyana, located at the end of the long and ardent valley of Barzrud in 
the central Karkas range. Its peculiar architecture and the distinctive attire of 
its women attract tourists from Tehran, where the younger generation of 
Abyana'i, typically affluent, live and work. Yet language has no role in this 
distinct identity. 


Cf. Yarshater, 2001, esp. p. 246; H. Borjian, 2005a; H. Borjian and M. Borjian, 2008. 
H. Borjian, 2005a. 

70 Krahnke, 1976, p. 62. Zhukovskii quotes the following popular rhyme from his rural 
informants who had added a fourth tongue to the famous trio: Lafz — lafz-e Arab ast; torki honar 
ast; fdrsi shakar ast; zaban-e ma [...] khar ast 'Language par excellence is Arabic; Turkish is (awork 
of) art; Persian is sugar; our language — crepitus podicis asini' (Zhukovskii, I, p. viii). 

71 Millspaugh, p. 256. 

78 H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 

Modernization and Language Shift 

While Iran as a whole underwent an enormous social and economic change in 
the past century, the adjustment has been even more profound in the back- 
ward communities speaking Central Dialects. The change began to be felt in 
the villages by the 1960s, when the Land Reforms broke the backbone of mil- 
lennia-old Persian feudalism and caused many former peasants to leave their 
home villages for cities to work at paying jobs. Another contemporaneous 
reform was the dissemination of literacy by means of a corps of young army 
recruits {Sepdh-e ddnesh) assigned to villages; in consequence, for the first 
time in history, Persian was introduced in a large scale to Median-speaking 
villagers at an early age. The older generation would have been excluded from 
this Persian 'invasion had the transistor radio not been made available to a 
majority of rural households at a low price by the end of the decade. Notwith- 
standing these facts, the outcome of modernization was still meager even a 
decade after the reforms had been introduced. When Karl Krahnke visited the 
area to document Central Dialects, the occupation of the villagers was still 
almost exclusively agriculture and carpet weaving, and the villages consisted 
of traditional mud-brick huts with little signs of modernization, including 
electricity. 72 

The rural conditions had altered beyond recognition when I began to visit 
the area in the late 1990s. Soon after the Revolution of 1979, men of the 
Rural Construction Corps arrived in virtually every village of the Central Dia- 
lect area and began to build roads, mosques, schools, and houses, brought 
electricity and telephone lines to the remotest of hamlets and set up a network 
of producer-and-consumer cooperatives. I came upon few households whose 
members were exclusively occupied in agriculture. Few younger males were 
not involved in a business or trade which did not in one way or another require 
routine trips to the nearby towns. As the weekend or a major holiday approaches, 
the traffic flow from cities to villages is noticeable. The enormous movement 
of sections of the rural population into towns and cities has contributed to the 
dispersal of rural communities and brought them into increased contact with 
the Persian language. There was hardly anyone in the villages I visited who 
could not speak Persian fluently. 

A powerful cause of language shift in the Central Dialect area has been 
compulsory mass education in Persian. It is true that Persian was the means of 
education even in olden times, yet formal education was largely unavailable in 
remoter villages and more so to women, who eventually transmit the native 

Krahnke, 1976, p. 19. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 79 

language to their children. In 1956, the literacy rate was as low as 15 percent 
for Isfahan province. With the enormous expansion of the educational system 
in the following decades, the literacy rate for the age group 6-39 approached 
98 percent in 2002. 73 Moreover, the statistics bear witness to the gradual taper- 
ing of the gender gap: girls' school enrollment increased from 19 percent in 
1953 to slightly more than half by the end of the century 74 In the post-revo- 
lutionary period, several colleges have been established in several townships, 
attracting students from all over the country and leading to greater diffusion 
of Persian at the expense of native dialects. 

While there has been no institutional support for Central Dialects, neither 
is there any indication that the use of these dialects has been discouraged by 
the government or by the speakers of Persian in general. Nevertheless, the vil- 
lagers in Kamandan east of Isfahan told me that their clergyman had warned 
them not to teach the vernacular to their children lest they never become pro- 
ficient in Persian. Whether this is a personal conviction of an individual mul- 
lah or part of the cultural policy advocated by the Shi'ite seminaries centered 
at Qom needs further investigation. 

Jewish and Zoroastrian Speakers 

Among the speakers of Central Dialects, religious minorities warrant particu- 
lar sociolinguistic attention for several reasons. Unlike the rural speakers of 
Central Dialects who otherwise have no distinct identity from their fellow 
countrymen, the religious minorities of Iran are both socially and officially 
distinguished from the Shi'ite mainstream. The Jewish speakers of the ver- 
naculars were, until recently, the only groups who had preserved the native 
Median in larger urban centers such as Isfahan, Kashan, and Hamadan. These 
Jewish communities, and the Zoroastrians of Yazd, have for centuries been 
living in Persian-speaking towns, and, consequently, they fall into a different 
social context from the rest of the speakers of Central Dialects who live by and 
large in relatively isolated villagers and townships. As the more liberal policies 
of the Pahlavi dynasty (1925-79) facilitated social integration and brought 
prosperity for the minority groups of the country, most of the Jews and Zoro- 
astrians moved to Tehran, which proved more hospitable and less prone to 
sectarian prejudices, as well as financially more attractive to these economi- 
cally dynamic minorities. Another wave of mass migration occurred during 

73 See M. Borjian and H. Borjian, 2007. 

74 Ibid. 

80 H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 

and after the Islamic Revolution of 1978-79, when religious minorities left for 
various countries and many eventually settled in North America. Having 
occurred within a generation or two, these migrations have meant a virtual 
end to the dialects, at least for the second generation of immigrants who have 
grown up in societies (Tehran or abroad) that did not encourage the use of the 

Jewish communities. Jewish communities are as deeply rooted in Persian society 
as the rest of Persians. Most urban centers of Persia used to have a Jewish quar- 
ter, typically by the side of the city's citadel, under the protection of the rulers 
who supported the Jewry for their expertise in commerce. There remains from 
medieval times a considerable body of Persian literature in Hebrew script 
known as Judo-Persian, 75 including some of the Biblical translation and com- 
mentaries in archaic forms of Persian. The Jewish residents of the central 
Persian towns generally preserved the native language that was lost otherwise 
to Persian. These include: (1) Hamadan and the nearby towns Nehavand, 
Malayer, and Tuysekan, and Arak, all geographically isolated from the con- 
tinuum of Central Dialects; (2) Kashan, Khomeyn, Golpayegan, and Isfahan — 
in these towns, too, the use of Median was exclusive to Jewish neighborhoods; 
(3) Khunsar, Delijan, and Mahallat, where Median was shared with the Mus- 
lim residents; and (4) Yazd and Kerman, where Median is also preserved 
by the Zoroastrians. Most of these longstanding Jewish communities were 
on the verge of disappearing already by 1970. The larger centers of Jewish 
population speaking a Central Dialect included Kerman and Yazd with about 
450 Jewish souls each, Hamadan with 800, and Isfahan with 2,500 to 3,000. 7S 
The larger part of the Jewish populace of these towns had already moved to 
Tehran, which hosted some 60,000 Jews at the time, or else they had immi- 
grated to Israel, a trend that has been continuing to this date. 

Among the Median-speaking Jewish communities, that of Isfahan has been 
by far the largest and one of the longest standing. 77 Its historical significance is 
implied by the fact that Isfahan was known in the early Islamic centuries as 
Johudestan/Yahudiya 'Jewish town,' referring to the large Jewish settlement on 
the site which eventually became the Jubara quarter in central Isfahan. The 

75 Judo-Persian is often confused with the Median dialects spoken by the Jews. 

76 Yarshater, 1974. 

The establishment of the Jewish colony of Isfahan is ascribed to those exiled by the Babylo- 
nian king Nebuchadnezzar, according to Talmudic tradition (s.v. "Isfahan," The Standard Jewish 
Encyclopaedia). The Middle Persian Shahristdnihd i Erdn states that the Sasanian king Yazdegerd I 
(r. 339-420) settled Jews in Isfahan at the request of his Jewish wife Shoshandukht. See Mark- 
wart, 1931, p. 19, par. 47. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 8 1 

Jewish population of the city remained sizeable throughout the Middle Ages, 
some 15,000 when the total population was about 100, 000. 78 From this 
historical backdrop remain some twenty synagogues in the old quarter of 
Jubara alone, as well as emdmzdda Esma'il, which is shared with Muslims, and 
the mausoleum of Pir-e Bakran near Isfahan, revered by Jewish pilgrims from 
throughout the country A rapid decline of population has been witnessed 
in the mid-twentieth century: of the 10,000-12,000 Jews who had lived in 
Isfahan in 1948, only about 2,500 remained in 1968, 79 and the number of 
functioning synagogues dropped to thirteen by 1961. 80 The current size of the 
community, estimated at about 2,000 souls by my local informants in recent 
years, may appear not radically different from the 1968 estimate; however, if 
population growth by natural birth is taken into account, the ongoing ele- 
ment of emigration, particularly to Israel, becomes more apparent. This pro- 
cess becomes even more evident in the number of operating synagogues, which 
has dropped to merely two or three in Jubara. On the other hand, the modern 
synagogue of Keter Davud in central Isfahan is the chief congregation of 
the community as it attracts, according to my informants, up to 500 people 
every Saturday morning. This in turn reflects the aerial shift in the Jewish 
population of the town: only a third of them still live in Jubara; the rest have 
moved to more affluent middle-class neighborhoods around the thoroughfare 
of Charbagh Avenue 81 and, accordingly, gradually assimilating into the gentile 

Indeed, social integration and education are prime factors in language shift 
among the fading Jewish community of Isfahan. Eager to avoid distinctive 
marks in public, the younger generations often consider their own dialect a 
feature reminiscent of their former inferior status. Traditionally peddlers and 
spinners living in ghettos, their condition improved substantially during the 
Pahlavi period, with many integrating into the mainstream socio-economic 
life and beginning to participate in virtually all walks of social life. In these 
developments, education and connection to the international Jewish move- 
ment have a role. The opening of the Alliance Israelite Universelle in 1901 was 
a turning point for modern education in the city. More Jewish schools were 
established in Isfahan by the ORT and Ozar ha-Torah in the first half of the 
twentieth century. Already in 1961, 150 pupils attended Jewish high school 

78 Sources on population of Isfahan and its Jewry are provided in H. Borjian, 1993. 
75 An estimate from the mid-1950s: 5,700 to 6,100, including 2,000 adults, 2,400 school 
age, and 1,500 preschool children; see 'Abedi, 1955, p. 220. 

80 "Isfahan," Encyclopaedia Judaica 9: 78-79. 

81 For the distribution of Jewish households among various neighborhoods of Isfahan, see the 
map in Shafaqi, 2003, p. 402. 

82 H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 

and 897 attended elementary school; other children attended non-Jewish 
schools, while there were about 50 Jews at the University of Isfahan. 82 

No longer operating in any domain of language use, the Jewish dialect of 
Isfahan is facing a dead-end fate. My aged informant told me that he uses Jidi 
but only occasionally, to exchange several sentences with a few men of his own 
age group. In the Persian Jewish community of New York City, there are still 
people who speak the dialect more or less fluently, but again, it rarely happens 
that they come across another fluent speaker to have a conversation in the 

Zoroastrian communities. The Zoroastrians of Persia may share their fate with 
the Jews in terms of incipient language loss, but not necessarily for the exact 
same causes and circumstances. Once scattered throughout Persia, the Zoro- 
astrian communities shrank considerably during the theocratic reign of the 
later Safavids, to the extent that, by the late nineteenth century, they were by 
and large limited to Yazd in central Persia and Kerman to its southeast. For 
many centuries, Yazd was the Zoroastrian stronghold with some two dozen 
villages and a quarter of the town itself populated by the followers of the 
"good religion." 83 The Zoroastrian population was estimated by European 
travelers of the early twentieth century to be about 5,000 in the town of Yazd, 
another 5,000 in its villages, more than 2,000 in Kerman, and 500 in the 
adjoining villages. 84 During the ensuing decades most Zoroastrians migrated 
to Tehran, the ever-expanding capital, which, according to an estimate, hosted 
19,000 of the total 25,000 Zoroastrians of Persia in 1975. 85 As successful 
entrepreneurs and professionals, Zoroastrians were in the forefront of the mass 
emigration during the Islamic Revolution. Many settled in North America, 
particularly in California and New York, where they have established their 
fire temples and endowments following their rich philanthropic traditions, 
often in collaboration with the Parsee co-religionists who have emigrated 
from India. 

The strong sense of pride among the Zoroastrian community has left its 
impression on the awareness of and attitude toward their native dialect. Zoro- 
astrians are regarded as the inheritors of one of the oldest religions and the 
glorious Sasanian traditions otherwise lost during the Arab invasion of the 
country in the seventh century of our era. The dialect spoken by the commu- 
nity is not usually percieved as merely one of Central Dialects as it really is, 

"Isfahan," Encyclopaedia Judaka 9: 78-79. 
Bailey, 1936. 

Vahman and Asatrian, 2002, pp. 13-15. 
Windfuhr, "Central Dialects." 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 83 

but as one of the oldest and purest of all living languages in the country. It is 
indeed one of the earliest collected by European travelers and the most exten- 
sively documented of all Central Dialects by the Iranians. Both the self-esteem 
and outside attention have probably been responsible for the positive view the 
Zoroastrians hold toward their idiom. To have an appropriate name for their 
dialect, they have replaced the label Gabri, on account of its negative connota- 
tion, 86 with Dari, which is in fact a prestigious medieval name for the Persian 
language; more recently the appellation Behdinan "the people of the Good 
Religion" has been extended to refer to the dialect as well. Although Yazdi has 
many distinct subdialects, the remarkable similarity between the Median dia- 
lects of Yazd and Kerman, and the intermingling of the Zoroastrians of two 
cities in Tehran and abroad has yielded a coine that allows all Persian Zoroas- 
trians to use it as a means of internal communication. In the fire temple of 
New York City casual communication is conducted in that vernacular. More- 
over, attempts has been made to establish classes for children in this dialect in 
order to keep the otherwise vanishing language alive. Whether or not these 
efforts can preserve the Zoroastrian dialect is yet hard to tell, as the Iranian 
Zoroastrian communities in diaspora generally consider Persian their princi- 
pal heritage. 


This essay claims not to be more than a general overview of the ongoing lan- 
guage shift in central Iran that raises some questions for future studies. Exam- 
ination of sources lead to the conclusion that the central-western Iranian 
Plateau, corresponding to the ancient Media Major, had not yet become Per- 
sianized to any large degree before the medieval period, but was inhabited by 
speakers of the dialects known today as theTatic and Central Dialect groups. 
When did the process of shift begin and how far did it advance throughout 
various stages of history? It is known but poorly how New Persian infiltrated 
Media beginning in the second millennium of the Common Era, though Per- 
sian did not fully assert itself until after the Safavids established their theo- 
cratic regime. As was the case with Turkish in Azerbaijan, a big shift to Persian 
might have taken place in the urban centers of central Persia in modern times 
due to the forces which still need to be established by examining historical 
sources for possible mass migrations, the role of trade and religion, state and 
local language policies, among others. 

86 The word gabri gawr had the original sense of 'man, Zoroastrian' but is used as 'infidel, fire 
worshipper' by Muslims. 

84 H. Borjian /Journal ofPersianate Studies 2 (2009) 62-87 

Many Central Dialects have been replaced by varieties of Persian, some in 
recent memory and for some others the shift came about centuries ago. While 
our understanding of the language situation in the remoter past is largely 
based on speculation, concrete evidence comes into view in the nineteenth 
century, thanks to the notes left by curious travelers and then by anthropolo- 
gists and linguists. Comparing older reports against more recent ones reveals 
the fact that Persian has been quite active in undermining the local dialects in 
many villages and towns of the Central Dialect area, perhaps more rapidly 
than ever before. While some geographical patterns of the language shift were 
identified in this paper, many questions remain — for example, the reason why 
the contemporary shift is far more evident in the northern and western parts 
than in the Isfahan district. A deeper analysis of the extent of communicative 
patterns, particularly the role of economy and the marketplace, may lead to 
better understanding of the causes and patterns of language shift. 

The causes of language shift within the modern setting include lack of eth- 
nic and tribal identity, lack of prestige due to low social and economic status 
of the speakers of Central Dialects, and finally the new and irreversible forces 
concomitant with modernization. In the past few decades, for the first time in 
history, transportation, urbanization, mass media, and public education have 
breached the relative isolation which had provided a haven for the survival of 
the language and other cultural traits unique to the speakers of Central Dia- 
lects. The small number of speakers of each speech community in the omni- 
presence of Persian certainly adds to the level of language endangerment. The 
urbanized Jewish and Zoroastrian speakers of Central Dialects began to be 
absorbed into the mainstream of the Persian-speaking middle class, another 
universal condition that generally leads to language shift. Mass emigration 
during and after the Islamic Revolution has been a trend specific to these 
speakers of Central Dialects. Generally speaking, Central Dialects are no lon- 
ger subject to gradual disappearance, which varied in scale relative to the his- 
torical conditions, geographical configuration of villages or social status of the 
speakers, as used to be the trend before the advent of modernization; we are 
now facing a condition that suggests progressive ousting, leading to the extinc- 
tion of these idioms within a generation or two, unless counteractive forces 
alter the equilibrium in favor of survival. 

What might these preservationist forces be? Some kind of institutional sup- 
port, such as state policy for maintaining endangered languages, is not likely 
to materialize in the foreseeable future, before it is too late. A more viable 
scenario would be a change of attitude among the speakers of the dialects 
towards their mother tongues. The population speaking these dialects, for- 
merly illiterate peasants or lower-class urban dwellers, are now skilled workers, 
educators, and professionals. While education can be held responsible as a 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 8 5 

catalyst in ousting dialects, it has also caused some speakers to be less indiffer- 
ent than their ancestors were towards their native culture, language included. 
This trend has been observed sporadically for some time now. A small group 
of former residents of Sedeh in Isfahan attempted to preserve the language, in 
part by publishing poetry 87 The fact that the Zoroastrians in diaspora have 
established classes to teach their tongue to their children is of great impor- 
tance, since the pulse of a language lies in its younger generation. There are 
local enthusiasts who have been collecting words, phrases, and poetry, and 
graduate students, mostly native speakers, have written scores of university 
projects on these dialects. These publications, notwithstanding their mediocre 
status from a scholarly perspective, have been on the rise in the last few decades, 
partly due to the cheap cost of print, something that may help at least some of 
the dialects eventually gain literary status, another positive sign for language 

Realistically speaking, however, these forces of preservation appear scanty 
when compared with the social realities mandating language shift. If mainte- 
nance of the highly endangered and moribund languages of central Iran is 
beyond practical help, the very last chance still remains to document what is 
left, and thus make a contribution to preserving the dialects, if only in archive 


Hasan Abedi, Esfahdn az lehdz-e ejtemai o eqtesddi, Tehran, 1955. 

Ruben Abrahamian, Dialectologie iranienne. Dialectes des Israelites de Hamadan et d'Ispahan et 

dialect des Baba Tahir, Paris, 1930. 
M. Adib Tusi, "Se ghazal-e esfahani az Owhadi-e maraghi," Nashriya-ye Ddneshkada-ye adabiydt-e 

Tabriz 15, 1963, pp. 387-400. 
Harold W. Bailey, "Iranian Studies. IV," BSO(A)S 7, 1935, pp. 769-78. 
Idem, "Yazdi," BSO(A)S 8, 1936, pp. 335-61. 

Habib Borjian, "Esfahan dar tarikh-e Iran," Rahdvard, ser. no. 33, 1993, pp. 74-95. 
Idem, "Tashih-e motun-e guyeshi: Divan-e Darvish Abbas-e gazi," Iranshenasi 17, 2005a, 

pp. 320-337. 
Idem, "Joghrafia-ye guyeshha-ye velayati-e Esfahan," Iranshenasi 17, 2005b, pp. 466-486. 
Idem, "Isfahan," in M.R.T. Dumper and B.E. Stanley, eds., Cities of the Middle East and North 

Africa: A Historical Encyclopedia, Santa Barbara, 2007a, pp. 147-149. 
Idem, "Isfahan xiii. Traditional Crafts," Encyclopaedia Iranica XIV, 2007b, pp. 48-55. 
Idem, "Isfahan xiv. Modern Economy and Industries," Encyclopaedia Iranica XIV, 2007c, 

pp. 55-67. 
Idem, "Isfahan xx. Geography of the Median Dialects of Isfahan," Encyclopaedia Iranica XIV, 

2007d, pp. 84-93. 

Krahnke, 1976, p. 62. 

86 H. Borjian /Journal ofPersianate Studies 2 (2009) 62-87 

Idem, "The Median-Parthian Group of the Northwest Iranian Languages," in H. Borjian, ed., 

Third Biennial Convention, Association for the Study ofPersianate Societies, Tbilisi, Georgia, 

8-11 June 2007. Abstracts and Presidential Address, Yerevan, 2007e, pp. 59-62. 
Idem, "Jarquya, District and Dialect," Encyclopaedia Iranica XIV, 2008a, pp. 582-588. 
Idem, "The Extinct Language of Gorgan: Its Sources and Origins," Journal of the American 

Oriental Society 128, 2008b, forthcoming. 
Idem, "When and How Did the Vernacular of Isfahan Shift from Median to Persian?" 2009, 

Habib Borjian and Maryam Borjian, "Twenty-five Mazandarani quatrains," Nawabi Memorial 

Volume, ed. M. Jaafari Dehaghi, Tehran, 2008, forthcoming. 
Maryam Borjian, "Bilingualism in Mazandaran: Peaceful Coexistence with Persian," Languages, 

Communities, and Education: A Volume of Graduate Student Research, New York: Teachers 

College, Columbia University Press, 2005, pp. 65-74. 
Maryam Borjian and Habib Borjian, "Isfahan xv. Education and Cultural Affairs," Encyclopaedia 

Iranica XIV, 2007, pp. 67-71. 
Arthur Christensen, Contribution a la dialectologie iranienne I, Copenhagen, 1930. 
Wilhelm Eilers and Ulrich Schapka, Westiranische Mundarten aus der Sammlung W Eilers, 3 

vols., Wiesbaden, 1976-88. 
Bahram Farahvashi, Vdzhandma-ye khuri, Tehran, 1976. 
Ronald W Ferrier, tr. and ed., A Journey to Persia: Jean Chardins Portrait of a Seventeenth-century 

Empire, London and New York, 1996. 
V Ivanow, "Two Dialects Spoken in the Central Persian Desert," JASB 21, 1926, pp. 405-31. 
Idem, "Notes on the Dialect of Khur and Mihrijan," Acta Orientalia 8, 1927, pp. 45-61. 
Idem, "The Gabri Dialect Spoken by the Zoroastrians of Persian," Rivista degli studi orientali 

16, 1934, pp. 31-97; 17, 1938, pp. 1-39; 18, 1939, pp. 1-59. 
Aqa Mir Sayyed Ali Jenab, al-Esfahdn, Isfahan, 1992. 
Karl J. Krahnke, Linguistic Relationships in Central Iran, Doctoral dissertation, University of 

Michigan, 1976. 
Ann K.S. Lambton, "The regulation of the waters of the Zayande Rud," BSOAS 9, 1937-39, 

pp. 663-673. 
Idem, Three Persian Dialects, London, 1938. 
Pierre Lecoq, Le dialecte de Sivand, Wiesbaden, 1979. 
Idem, "Les dialects du centre de llran," in R. Schmitt, ed., Compendium. Linguarum Iranicarum, 

Wiesbaden, 1989, pp. 313-326. 
Idem, Recherches sur les dialectes kermaniens (Iran central), Acta Iranica 39, Belgium, 2002. 
D.L.R. Lorimer, "Notes on the Gabri Dialect of Modern Persian," JRAS, 1916, pp. 423-89. 
Idem, "Is there a Gabri Dialect of Modern Persian?" JRAS, 1928, pp. 287-319. 
David Neil MacKenzie, "The origins of Kurdish," Transactions of the Philological Society, 1961, 

pp. 68-86. 
Mohammad-Reza Majidi, Guyeshhd-ye pirdmun-e Kdshdn o Mahalldt, Tehran, 1975. 
Oscar Mann and K. Hadank, Kurdisch-persische Forschungen 3/1. Die Mundarten von Chunsar, 

Mahallat, Natanz, Nayin, Samndn, Sivand und So-Kohrud, Berlin and Leipzig, 1926. 
J. Markwart, A catalogue of the provincial capitals of Eranshahr: Pahlavi text, version and com- 
mentary, ed. G. Messina, Rome, 1931. 
A.C. Millspaugh, The American Task in Persia, New York, 1925. 
Haideh Sahim, "Languages and dialects of the Jews of Iran and Afghanistan," in Human Sarshar, 

ed., Esthers Children: A Portrait of Iranian Jews, Beverly Hills, Cal., 2002, pp. 283-294. 
Rudiger Schmitt, "Andere altiranische Dialekte," in R. Schmitt, ed., Compendium Linguarum 

Iranicarum, Wiesbaden, 1989, pp. 86-94. 
AH Shafi'i Nikabadi, Garkuya: Sarzamin-i ndshendkhta bar kardn-e kavir, Isfahan, 1997. 

H. Borjian /Journal of Persianate Studies 2 (2009) 62-87 87 

Sirus Shafaqi, Joghrdfid-ye Esfahdn, Isfahan, 2003. 

Maxim Siroux, Anciennes vols et monuments routiers de la region d'Isfahan, Cairo, 1971. 

Shendsdyi-e guyeshhd-ye Iran, 4 vols., Tehran, 1983-2000. 

Donald Stilo, "Isfahan xxi. Provincial Dialects," Encyclopaedia Iranica XIV, 2007, pp. 93-112. 

Idem, "A Description of the Northwest Iranian Project at the Max Planck Institute for 

Evolutionary Anthropology," at; cited on 

Nov. 1, 2007. 
Ahmad Tafazzoli, "Ettela'at-i darbara-ye lahja-ye pishin-e Esfahan," Ndma-ye Minovi, Tehran, 

1971, pp. 85-103. 
Y. Thamara, "Mo'arrefi-ye do tarh-e melli-e guyesh-shenasi," in H. Reza'i Baghbidi, ed., Majmu'a- 

maqdldt-e nokhostin hamandishi-e guyesh-shenasi dar Iran, Tehran, 2002, pp. 135-142. 
F. Vahman and G. Asatrian, Notes on the Language and Ethnography of the Zoroastrians ofYazd, 

Copenhagen, 2002. 
Gemot L. Windfuhr, review of Ehsan Yarshater's A Grammar of Southern Tati Dialects, in JAOS 

92, 1972, pp. 370-374. 
Idem, "Isoglosses: A Sketch on Persians and Parthians, Kurds and Medes," in Acta Iranica. 

Monumentum H.S. Nyberg II, Acta Iranica 4-7, Tehran and Liege: Brill, 1975, pp. 457- 

Idem, "Western Iranian Dialects," in R. Schmitt, ed., Compendium Linguarum Iranicarum, 

Wiesbaden, 1989, pp. 294-294. 
Idem, "Central Dialects," Encyclopaedia Iranica V, pp. 242-252. 
E. Yarshater, A Grammar of Southern Tati Dialects. Median Dialect Studies I, The Hague, 

Idem, "The Jewish communities of Persia and their dialects," Melanges Jean de Menasce, Louvain, 

1974, pp. 455-466. 
Idem, "Azerbaijan vii. The Iranian language of Azerbaijan," Encyclopaedia Iranica III, 

pp. 238-245. 
Idem, "A Peasant Marriage: A Poem in Chali by Mohammad-Baqer Ameli," Studia Iranica 

30, 2001, pp. 245-289. 
V.A. Zhukovskii, Materiali dlya izuchesniya persidskikh' narechiy, 3 vols., St. Petersburg, 1888- 


Copyright of Journal of Pcrsianatc Studies is the property of Brill Academic Publishers and its 
content may not be copied or emailed to multiple sites or posted to a listscrv without the copyright 
holder's express written permission. However, users may print, download, or email articles for 
individual use.